[Bug 3813] Editorial: correct misuse of 'lexical mapping'

http://www.w3.org/Bugs/Public/show_bug.cgi?id=3813





------- Comment #1 from davep@iit.edu  2006-10-09 19:29 -------
(In reply to comment #0)

> Section 3.3.5.1 has
> 
>     Note: As explained below, the lexical mapping of the float
>     value notANumber is 'NaN'.  Accordingly, in English text we
>     generally use 'NaN' to refer to that value.

> The lexical mapping of a type is a function from the lexical
> space to the value space; that term makes no sense in the
> sentences just given.  What is meant is the pre-image of
> notANumber.

Since the special value notANumber is in the value space of float and decimal,
'NaN' is in the lexical space, and the lexical mapping is a function which,
among other things, maps 'NaN' to notANumber,
I'd say the Note is correct as stated.

> Section 3.3.4.1 on decimal has a better wording:
> 
>     Note: As explained below, the lexical representation of the
>     precisionDecimal value object whose numericalValue is
>     notANumber is 'NaN'.  Accordingly, in English text we use
>     'NaN' to refer to that value.

In the case of precisionDecimal, the values in the value space are three-tuples
with named coordinates; one of the coordinates is named "numericalValue".  When
the numericalValue coordinate has the value notANumber, the arithmeticPrecision
and sign coordinates must both have the value absent.  So the lexical mapping
maps 'NaN' to that three-tuple.  In the case of float and double, the constant
is the value, rather than being one coordinate of a three-tuple value. In all
three cases, the lexical mapping maps 'NaN' to the corresponding value in the
value space.

Received on Monday, 9 October 2006 19:29:53 UTC